ENERGY BALANCE OF TRUMPETER SWANS AT STOPOVER AREAS DURING SPRING MIGRATION

NORTHWESTERN NATURALIST 85:104 110 WINTER 2004 ENERGY BALANCE OF TRUMPETER SWANS AT STOPOVER AREAS DURING SPRING MIGRATION JALENE MLAMONTAGNE 1,ROBERT MR BARCLAY, AND LELAND JJACKSON Ecology Division, Department of Biological Sciences, University of Calgary, 2500 University Drive NW, Calgary, Alberta, T2N 1N4, Canada; jalene.lamontagne@ualberta.ca ABSTRACT We investigated whether trumpeter swans (Cygnus buccinator) gain energy from forage in excess of estimated daily demands while present in a spring migration stopover area west of Calgary, Alberta. We estimated energy budgets from activity time budgets of 443 individual focal-animal observations. We conducted exclosure experiments in ponds used by trumpeter swans to estimate consumption of Potamogeton pectinatus tubers and rhizomes, and we determined metabolizable energy of forage to assess energy intake. Our conservative estimate of forage removal indicated that energy demands were similar to gains within migration stopover areas. Therefore, trumpeter swans likely have a positive energy balance that is available for further migration and reproduction. We suggest that spring migration stopover areas could play a critical role in the further conservation of this species. Key words: trumpeter swan, Cygnus buccinator, energetics, spring migration, Alberta Many studies have been conducted on the breeding and wintering ecology of trumpeter swans (Cygnus buccinator) (for example, Shea 1979; Holton 1982; Maj 1983; Squires 1991), yet scant information exists regarding their spring migration ecology (LaMontagne and others 2001; LaMontagne and others 2003a, 2003b). Stopover areas may provide essential energy reserves necessary for successful reproduction; therefore, such areas may be critical to the continued recovery of trumpeter swans in North America. The majority of trumpeter swans in the Rocky Mountain population migrate approximately 1400 km in spring from overwintering areas in the tri-state region of Montana, Idaho, and Wyoming, along a narrow corridor along the east front of the Rocky Mountains to Grande Prairie, Alberta, northeastern British Columbia, or the Yukon to breed (Mackay 1978; Drewien and Shea 2003). They arrive in April and early May when the ponds are still iced over and begin nest building shortly after arrival (Holton 1982; Gale and others 1987). In these areas there are little exogenous resources available to swans during ovulation and initiation of incubation (Holton 1982); therefore, the 1 Present address: Department of Biological Sciences, University of Alberta, Edmonton, Alberta, T6G 2E9, Canada. swans are likely dependent on endogenous lipid and protein reserves acquired during spring migration. A positive correlation exists between female condition upon arrival at breeding grounds and reproductive output in geese (Ankney and MacInnes 1978; Raveling 1979; Davies and Cooke 1983). Likewise, it has been assumed trumpeter swans make energy gains in spring migration stopover areas and that these are later allocated to reproduction (Mitchell 1994). Our goal in this study was to estimate the energetic demands of trumpeter swans at a spring migration stopover area. We used time budgets of diurnal activity to relate energetic gains through consumption of Potamogeton pectinatus rhizomes and tubers to energetic costs in order to assess whether trumpeter swans leave spring migration stopover areas with a net energy gain that can later be allocated to reproduction. METHODS Our study area was a spring-migration stopover area located approximately 50 km W of Calgary, Alberta (51 05 N, 114 30 W to 51 09 N, 114 42 W). This site is in the foothills of the Rocky Mountains, and ranching is the main land practice. There are many ponds in the area, some of which are used every spring 104

WINTER 2004 LAMONTAGNE AND OTHERS: TRUMPETER SWAN ENERGETICS 105 by trumpeter swans (LaMontagne and others 2003a). Swans were present in our study area from 7 April to 21 May 1999. During this time we counted the number of individuals present once each day on 13 ponds in the area and we made 443 individual 10-min observations using focal-sampling techniques on 4 study ponds. The 13 ponds ranged in size from 0.5 to 9.2 ha, with submerged macrophyte cover of 0 to 84%, ph values between 7.97 and 9.63, salinity of 0.31 to 1.11 ppt, and elevations from 1250 to 1332 m (see also LaMontagne and others 2003a). During each observation we continuously recorded the time the individual spent in 6 activities (foraging, preening, resting, sleeping, flying, and swimming) and whether the individual observed was an adult (white plumage) or a juvenile (gray plumage). We conducted observations between 0600 and 2100. We observed 10 randomly selected individuals per pond, per day, with no adjustments for the number of adults and juveniles on the pond. If 10 swans were present on a pond, all swans were observed. We assumed that individual observations were independent, as family groups were typically not discernable (see also La- Montagne and others 2003b). We used the average time budget of swans on 4 study ponds (see below) and estimated an energy budget by assigning to each behavior a multiplier that was a function of the basal metabolic mate (BMR) (Dolnik 1980, cited in Kritsov and Mineyev 1991). We combined data from adults (n 361) and cygnets (n 82) as there was no significant difference in their time budgets (MANOVA, P 0.05). This is consistent with the findings of Squires and Anderson (1997) during spring in the Yellowstone area. The BMR of trumpeter swans was calculated as: BMR (in kj) 307.7W 0.73, where W is mass in kilograms (Aschoff and Pohl 1970). An average mass of 10.8 kg was used for adults and cygnets based on the relative abundance of adults and cygnets counted on study ponds (LaMontagne, unpubl. data) and the winter body mass of female and male adults and cygnets (Drewien and Bouffard 1994), assuming a 50:50 sex ratio. Trumpeter swans are active at night with no differences between diurnal and nocturnal time budgets (Squires and Anderson 1997), so we extrapolated our diurnal time budgets over the 24-h period. We estimated trumpeter swan energy requirements for a single day by multiplying the proportion of time allocated to each behavior and the value of the behavior relative to BMR. We summed over all behaviors and multiplied by the BMR. We then estimated the total amount of forage required by all trumpeter swans to satisfy their daily energetic requirements. We estimated the total energy requirements of all swans present on ponds that they used consistently and divided by the average energy content of a dry gram of P. pectinatus tubers and rhizomes. We studied 4 consistently used ponds ( 100 total swan-days/y in both 1999 and 2000) and all were dominated by P. pectinatus. In 2 of these ponds, Sibbald and Sibbald East, we used exclosures to assess forage availability and the consumption of P. pectinatus tubers and rhizomes by trumpeter swans. We distributed 5 rebar and plastic-mesh exclosures (60 cm 60 cm each) roughly evenly on the surface of the frozen sediment prior to the swans arrival in each pond in 1999. After all trumpeter swans had left the study area, we randomly sub-sampled the sediment within each exclosed area 6 times with an Eckman grab (15 cm 15 cm 15 cm). We also selected additional sites in these 2 ponds where trumpeter swan foraging had been observed and took 6 Eckman samples from each of five 0.6-m 2 areas. We avoided sampling in areas where feeding craters were created by swans (up to 1 m diameter and 30 cm deep; Shea 1979), as there was little sediment in these areas to sample. In Beaver and Jumping Pound, the other 2 study ponds, placement of exclosures prior to the arrival of trumpeter swans was not logistically possible; therefore, we estimated tuber and rhizome availability in autumn 1999 using Eckman samples in the same patterns as noted above. We back-calculated spring rhizome and tuber biomass using the difference in mean tuber and rhizome biomass from Eckman samples taken in autumn 1999 and spring 1999 from exclosed areas of Sibbald and Sibbald East, as spring trumpeter swan foraging does not have a significant negative carry-over effect on summer P. pectinatus biomass (LaMontagne and others 2003b). We also took Eckman samples in the spring of 1999 from 9 additional ponds that we classified as variably used or unused. Eckman samples were sorted in the lab using a 4-mm sieve, and tubers and rhizomes were dried at 60 C for 48 h. We

106 NORTHWESTERN NATURALIST 85(3) 1971), suggesting we are not making a large leap of faith. However, this hypothesis can be tested as data become available. RESULTS FIGURE 1. Daily counts of trumpeter swans on 13 study ponds in the Calgary stopover area during the 1999 spring migration. calculated the total dry mass of tubers and rhizomes available to trumpeter swans as the total dry mass in the pond area 1 m deep, because trumpeter swans can only reach about 100 cm below the surface of the water surface to forage (Scott 1972; Holton 1982). Three 1-g samples each of tubers and rhizomes were analyzed for gross energy and metabolizable energy. Metabolizable energy equals (gross energy (fecal urinary energy)) (Alisauskas and Ankney 1992) and is considered to be the most meaningful expression of dietary content (Burton and others 1979). Gross energy and metabolizable energy were calculated from crude protein, crude fat, and crude fiber measurements. As calculations were performed in the lab, poultry was the most relevant bird for which values of metabolizable energy were available. Estimates for mallards (Anas platyrhynchos) and poultry have similar values of metabolizable energy (Sugden We counted 1150 trumpeter swan-days (a swan-day is equal to a single swan present for a single day) on our 4 study ponds and a total of 1300 swan-days on all 13 ponds in our study area during the 1999 spring migration (Fig. 1). The peak number of trumpeter swans in our study area was 165 on 14 April. The dominant activity on the 4 consistently used ponds was foraging, followed by swimming, sleeping, and preening. Resting and flying were uncommon during observations (Table 1; see also LaMontagne and others 2001). The BMR of a 10.8 kg bird is 1748 kj/day. By factoring in the daily activities of trumpeter swans, we estimated the daily energy requirement of an individual swan as 2578 kj/day (Table 1). The total energy requirement of the 1150 trumpeter swans on the four consistently used ponds in the stopover area west of Calgary was estimated to be 2,964,700 kj. The relative abundance of tubers and rhizomes in Eckman samples prior to trumpeter swan foraging was 6.1 % and 93.9 %, respectively (Table 2). Tubers had a significantly higher metabolizable energy content than rhizomes (t 27.25, df 4, P 0.001), but gross energy did not differ significantly (t 0.56, df 4, P 0.05). The average metabolizable energy content of available forage, based on the TABLE 1. Time-budgets and energy requirements for trumpeter swans on four study ponds west of Calgary in spring, 1999, based on the average mass of a trumpeter swan being 10.8 kg. Energy value of activities relative to basal metabolic rate (BMR) are from Dolnik (1980; cited in Kritsov and Mineyev 1991). Behavior Forage Preen Rest Sleep Fly Swim Proportion of time activity performed 0.543 0.135 0.023 0.143 0.001 0.155 Value of activity ( BMR) 1.6 1.3 1.12 1.0 14.0 1.6 Daily energy requirement ( BMR) 0.869 0.176 0.026 0.143 0.014 0.248 Total daily energetic requirement 1.475 BMR 1.475 1748 kj/day 2578 kj/day Total energy requirements of all swans: 2578 kj/day 1150 swan-days 2,964,700 kj

WINTER 2004 LAMONTAGNE AND OTHERS: TRUMPETER SWAN ENERGETICS 107 TABLE 2. Mean (standard error) energy content and composition of tubers and rhizomes collected from Sibbald, Sibald East, and Beaver pond west of Calgary. Three 1-g samples of tubers and rhizomes were analyzed, except for ash where only one sample of each was analyzed. Gross energy (kj/g) ns Metabolizable energy (kj/g)* Ash (%) Crude protein (%)* Crude fibre (%)* Crude fat (%)* Relative proportions in Eckman samples prior to swan foraging * P 0.05 from t-tests. ns P 0.05. Tuber 17.66 (0.01) 13.29 (0.04) 4.6 10.9 (0.2) 0.5 (0.2) 1.4 (0.1) 0.061 Rhizome 17.61 (0.07) 11.77 (0.04) 13.7 19.1 (0.9) 7.7 (0.4) 4.5 (0.1) 0.939 relative abundance of tubers and rhizomes in Eckman samples was 11.86 kj/g. Therefore, the total dry mass of forage required to meet the energy requirements of 1150 trumpeter swansdays in the 4 consistently used ponds in 1999 was about 250 kg. The total dry mass of forage available per pond was highest in Beaver followed by Jumping Pound, and then Sibbald and Sibbald East, which had similar amounts of total available forage (Table 3). The total dry mass of forage available to trumpeter swans to a depth of 15 cm in the sediment was 360 kg, or 38.5 kg/ha. After the swans continued their migration, the total dry mass of tubers and rhizomes remaining as estimated from the Eckman samples was 155 kg, or 16.6 kg/ha. Subtracting these values suggests that the swans removed an estimated 205 kg of forage. Thus, an estimated 57% of the tubers and rhizomes available were consumed by trumpeter swans in the spring of 1999, only 45 kg less than the estimated expected consumption based on energetic requirements. The percentage of tubers and rhizomes removed was highest for Jumping Pound and Beaver ponds, which also had the most swandays of use, while Sibbald and Sibbald East had lower swan use and a lower percent removal of tubers and rhizomes (Table 3). For the other 9 ponds in our study area that were not classified as consistently used, there was a total dry mass of only 190 kg of forage available in 15.8 ha, or only 12.0 kg/ha. DISCUSSION The estimated total amount of forage required by 1150 trumpeter swans-days for daily activities (250 kg), and the estimated amount of forage consumed in the 4 consistently used ponds (205 kg) was similar, suggesting that the swans just meet their energy requirements at migration stopover areas, based on our calculations. However, our estimate of the amount of forage removed was likely conservative, and we therefore suggest that trumpeter swans gain energy reserves in this migration stopover area. First, trumpeter swans can create craters up to 30 cm deep and 1 m across by scratching at the sediment to remove tubers (Shea 1979), and we avoided sampling in craters. Therefore, TABLE 3. Total dry mass (rhizomes plus tubers) available and remaining in consistently used ponds west of Calgary after trumpeter swans departed. Pond Jumping Pound a Beaver a Sibbald b Sibbald East b Total Area of pond 1 m deep (ha) 3.543 2.850 1.399 1.548 9.340 a Available forage estimates adjusted for decomposition. b Available forage estimated using exclosures. Total dry mass (kg) Available 90.12 155.92 59.28 54.58 359.90 Remaining 21.00 47.29 46.22 40.59 155.10 Percent removed 76.7 70.0 22.0 25.6 56.9 Total 1999 swan-days 411 320 223 196 1150

108 NORTHWESTERN NATURALIST 85(3) there were areas of the ponds where trumpeter swans may have more completely depleted tubers and rhizomes that we did not factor into our estimates of forage removal. Second, we could only sample the top 15 cm of the sediment, although trumpeter swans may forage to depths below 15 cm if the water depth is less than about 85 cm (Holton 1982) and larger tubers occur between 10 to 20 cm depth (Anderson and Low 1976). In combination, these factors mean that it is very likely that trumpeter swans gain not only the energy required to meet daily requirements but also energy to store for continued migration and/or breeding, although trumpeter swans moving through the study area very late may not be nesting. Although cygnets do not reproduce, they suffer higher overwinter mortality than older swans and those that survive may be in poor body condition (R Shea, Trumpeter Swan Society, Maple Plain, MN, pers. comm.). Therefore, cygnets may use excess energy they gain to enhance body condition and perhaps also for continued growth. The metabolizable energy of tubers and rhizomes from our study ponds was similar to that found in previous studies (McKelvey 1985; Squires 1991). In our study, however, rhizomes were much more abundant than tubers. Rhizomes have slightly less metabolizable energy than tubers, but they are higher in ash, crude fiber, crude fat, and crude protein (Table 2). Ash and fiber decreases the digestibility of plants (Muztar and others 1977; Thomas and Prevett 1980) and may explain why the metabolizable energy of rhizomes was lower than for tubers even though their gross energy contents were similar. Rhizomes had a significantly higher proportion of fat and protein compared to tubers (Table 2). Fat fuels flight during migration, can be carried without storing extra water, and contains more than twice the energy content of proteins and carbohydrates (Alerstam 1993). Endogenous fat and protein are used for egg synthesis, and food selection by waterfowl is influenced by the need to satisfy nutrient requirements of reproduction (Krapu and Reinecke 1992). Fat reserves possessed by females at the onset of breeding are an important determinant of clutch size in waterfowl (Ankey and MacInnes 1978; Krapu 1981). Therefore, the relatively high protein and fat content of rhizomes may meet nutritional demands of trumpeter swans better than tubers, although rhizomes have rarely been identified as a food source for trumpeter swans (but see McKelvey 1985), while tubers have been identified as a preferred forage (for example, Squires 1991). Given the high density of foraging trumpeter swans on our study ponds, the long-term sustainability of the macrophyte food base might be questionable. The timing of herbivory with respect to the growing season and the level of tissue consumption may have important implications for macrophyte abundance and community composition in subsequent growing seasons, including a shift in community composition with a reduction in favored macrophyte species and reduced macrophyte biomass (Hampton 1981; Lodge 1991; Perrow and others 1997; Mitchell and Perrow 1998). The future productivity of macrophytes would not be directly affected by foraging on aboveground parts in autumn when the plants are senescing (Kiørboe 1980). However, trumpeter swans arrive in migratory stopover areas in southern Alberta while ice still covers a large proportion of the ponds and macrophyte growth has not yet begun. The plants from the previous year have senesced; hence, the primary food source available for trumpeter swans in April is tubers and rhizomes. Consumption of these overwintering structures removes the future growth potential of that plant (Mitchell and Perrow 1998). However, in some areas P. pectinatus has flourished despite heavy grazing by waterfowl for 20 years (Kantrud 1986). Near Calgary, the consistently used ponds are dominated by P. pectinatus, and yearly variation in the timing and pattern of ice break up relative to the swans arrival may provide macrophytes with refuges from foraging swans during some years (LaMontagne and others 2003b). The pond area 1 m deep may also serve as a source for seeds, tubers, or rhizomes. Furthermore, 43% of available tubers and rhizomes remained after trumpeter swans had left the area, and peak summer macrophyte biomass does not differ significantly between areas where swans foraged and where they had been excluded (LaMontagne and others 2003b). Therefore, it is likely that the macrophyte community is sustainable. The role and importance of spring migration

WINTER 2004 LAMONTAGNE AND OTHERS: TRUMPETER SWAN ENERGETICS 109 stopover habitats for trumpeter swans has generally been overlooked. Our data suggest that trumpeter swans experience net energy gains on spring stopovers. Further research should include a sample of more stopover areas along the spring migration route to examine if activity budgets and energy balance of trumpeter swans change as they move northward from overwintering areas to breeding sites. Monitoring trumpeter swan migration with satellite collars could be used to more precisely follow migration paths and identify key stopover areas. Also, very little is known about habitat selection, activity budgets, and resource consumption of trumpeter swans in stopover areas during fall migration. ACKOWLEDGMENTS We thank A Dennis, C Solohub, and C Friedrich for help in the field and G Beyersbergen (Canadian Wildlife Service), D Brown (Alberta Environmental Protection), and L Hills for help identifying study ponds. This research was supported by Natural Sciences and Engineering Research Council awards to R Barclay and L Jackson, Challenge Grants in Biodiversity, a University of Calgary Thesis Research Grant, and an Ellis Bird Farm Graduate Scholarship to J LaMontagne. LITERATURE CITED ALERSTAM T. 1993. Bird migration. Cambridge, UK: Cambridge University Press. 428 p. ALISAUSKAS RT, ANKNEY CD. 1992. The cost of egg laying and its relationship to nutrient reserves in waterfowl. In: Batt BDJ, Afton AD, Anderson MG, Ankney CD, Johnson DH, Kadlec JA, Krapu GL, editors. Ecology and management of breeding waterfowl. Minneapolis, MN: University of Minnesota Press. p 30 60. ANDERSON MG, LOW JB. 1976. Use of sago pondweed by waterfowl on the Delta Marsh, Manitoba. Journal of Wildlife Management 40:233 242. ANKNEY CD, MACINNES CD. 1978. Nutrient reserves and reproductive performance of female lesser snow geese. Auk 95:459 471. ASCHOFF J, POHL H. 1970. Rhythmic variations in energy metabolism. Federation of American Societies for Experimental Biology Journal 29:1541 1552. BURTON BA, HUDSON RJ, BRAGG DD. 1979. Efficiency of utilization of bulrush rhizomes by lesser snow geese. Journal of Wildlife Management 43:728 735. DAVIES JC, COOKE F. 1983. Annual nesting productivity in snow geese: prairie droughts and arctic springs. Journal of Wildlife Management 47:291 296. DOLNIK VR. 1980. Coefficients for calculating energy expenditure by free-living birds from chronometry data of their activity. Ornitologia, part 15:63 74. (in Russian). DREWIEN RC, BOUFFARD SH. 1994. Winter body mass and measurements of trumpeter swans Cygnus buccinator. Wildfowl 45:22 32. DREWIEN R, SHEA R. 2003. Restoring severed migratory patterns of Rocky Mountain trumpeter swans and reconnection with essential wintering areas. http://www.y2y.net/science/grants/03 drewien. asp. Revised May 2004, accessed 19 May 2004. GALE RS, GARTON EO, BALL IJ. 1987. The history, ecology, and management of the Rocky Mountain population of trumpeter swans. Missoula, MT: US Fish and Wildlife Service, Cooperative Wildlife Research Unit. 314 p. HAMPTON PD. 1981. The wintering and nesting behavior of the trumpeter swan [thesis]. Missoula, MT: University of Montana. 185 p. HOLTON GR. 1982. Habitat use by trumpeter swans in the Grande Prairie region of Alberta [thesis]. Calgary, AB: University of Calgary. 164 p. KANTRUD HA. 1986. Sago pondweed (Potamogeton pectinatus L.): a literature review. Jamestown, ND: US Fish and Wildlife Service Resource Publication 176. 89 p. KIøRBOE T. 1980. Distribution and production of submerged macrophytes in Tipper Grund (Ringkøbing Fjord, Denmark), and the impact of waterfowl grazing. Journal of Applied Ecology 17:675 687. KRAPU GL. 1981. The role of nutrient reserves in mallard reproduction. Auk 98:29 38. KRAPU GL, REINECKE KJ. 1992. Foraging ecology and nutrition. In: Batt BDJ, Afton AD, Anderson MG, Ankney CD, Johnson DH, Kadlec JA, Krapu GL, editors. Ecology and management of breeding waterfowl. Minneapolis, MN: University of Minnesota Press. p 1 29. KRITSOV SK, MINEYEV YN. 1991. Energy budgets of whooper and Bewick s swans. Wildfowl, Supplement No. 1:319 321. LAMONTAGNE JM, BARCLAY RMR, JACKSON LJ. 2001. Trumpeter swan behaviour at spring-migration stopover areas in southern Alberta. Canadian Journal of Zoology 79:2036 2042. LAMONTAGNE JM, JACKSON LJ, BARCLAY RMR. 2003a. Characteristics of ponds used by trumpeter swans in a spring migration stopover area. Canadian Journal of Zoology 81:1791 1798. LAMONTAGNE JM, JACKSON LJ, BARCLAY RMR. 2003b. Compensatory growth responses of Potamogeton pectinatus to foraging by migrating trumpeter swans in spring stop over areas. Aquatic Botany 76:235 244.

110 NORTHWESTERN NATURALIST 85(3) LODGE DM. 1991. Herbivory on freshwater macrophytes. Aquatic Botany 41:195 224. MACKAY RH. 1978. Status report on trumpeter swan (Olor buccinator) in Canada. Status reports on endangered wildlife in Canada, vol. 1. Ottawa, ON: Canadian Wildlife Service, Committee on the Status of Endangered Wildlife in Canada. 23 p. MAJ ME. 1983. Analysis of trumpeter swan habitat on the Targhee National Forest of Idaho and Wyoming [thesis]. Bozeman, MT: Montana State University. 102 p. MCKELVEY R. 1985. The metabolizable energy of chicken scratch, the rhizomes of Carex lyngbei, and timothy grass to swans. Ottawa, ON: Canadian Wildlife Service. Progress Note no. 152. 3 p. MITCHELL CD. 1994. Trumpeter swan (Cygnus buccinator). In: Poole A, Gill F, editors. The birds of North America, No.105. Washington DC: Academy of Natural Sciences and Philadelphia, PA: American Ornithologists Union. p 1 24. MITCHELL SF, PERROW MR. 1998. Interactions between grazing birds and macrophytes. In: Jeppesen E, Sondergaard M, Sondergaard M, Christoffersen K, editors. The structuring role of submerged macrophytes in lakes. New York, NY: Springer-Verlag Inc. p 175 195. MUZTAR AJ, SLINGER SJ, BURTON JH. 1977. Metabolizable energy content of freshwater plants in chickens and ducks. Poultry Science 56:1893 1899. PERROW MR, SCHUTTEN JH, HOWES JR, HOLZER T, MADGWICK FJ, JOWITT AJD. 1997. Interactions between coot (Fulica atra) and submerged macrophytes: the role of birds in the restoration process. Hydrobiologia 342/343:241 255. RAVELING DG. 1979. The annual cycle of body composition of Canada geese with special reference to control of reproduction. Auk 96:234 252. SCOTT P. 1972. The swans. London, UK: Michael Joseph Ltd. 242 p. SHEA RE. 1979. The ecology of trumpeter swan in Yellowstone National Park and vicinity [thesis]. Missoula, MT: University of Montana. 132 p. SQUIRES JR. 1991. Trumpeter swan food habits, forage processing, activities, and habitat use [dissertation]. Laramie, WY: University of Wyoming. 213 p. SQUIRES JR., ANDERSON SH. 1997. Changes in trumpeter swan (Cygnus buccinator) activities from winter to spring in the Greater Yellowstone area. American Midland Naturalist 138:208 214. SUGDEN LG. 1971. Metabolizable energy of small grains for mallards. Journal of Wildlife Management 35:781 785. THOMAS VG, PREVETT JP. 1980. The nutritional value of sea-arrow grasses at James Bay. Journal of Wildlife Management 44:830 836. Submitted 9 May 2002, accepted 25 June 2004. Corresponding Editor: SR Johnson.